Scalable and Energy-Efficient Multimedia Content Sharing over New Generation Computing and Communication Platforms

通过新一代计算和通信平台实现可扩展且节能的多媒体内容共享

• 批准号: RGPIN-2014-04765
• 负责人: Liu, Jiangchuan
• 金额: $ 3.35万
• 依托单位: Simon Fraser University
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2017
• 资助国家: 加拿大
• 起止时间: 2017-01-01 至 2018-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=634301
• 关键词: Scalable; Energy; Efficient; Multimedia; Content

In the past decade, we have witnessed the fast evolution toward a new generation of social, mobile and cloud computing for networked multimedia processing and sharing. Today, high-definition videos and even 3D/multiview videos can be readily captured and browsed by personal computing devices, and conveniently processed and stored with remote cloud resources. The users are now actively engaged to be part of a social ecosystem, rather than passively receiving media content. The revolution is being driven further by the deep penetration of 3G/4G wireless networks and smart mobile devices that are seamlessly integrated with online social networking and media sharing services. Despite the increasingly abundant bandwidth and computation resources, the ever increasing data volume of user generated media content and the boundless coverage of socialized sharing have presented unprecedented challenges to both content and network service providers. The highly diversified media content types, origins, distribution channels further impose complex interactions among the different components in a networked sharing system. We thus ask the following critical question: how can the massive media content be efficiently generated and distributed among massive clients with the latest social, mobile, and cloud computing and communication platforms? Besides scalability, energy has become a critical concern with these new generation computation and communication platforms, too. On the mobile terminal side, the highly constrained battery reservoir has long been a bottleneck in service time between recharges; on the cloud side, the datacenters are known to be power-hungry. The computation- and bandwidth-intensive multimedia services that extensively use CPU, GPU, storage, and network resources certainly make the situation worse. Is has been predicted that multimedia energy consumption is rising at 16% per year, accounting for half of the annual operating expenditure of these services. To achieve the ultimate goals of scalable and energy-efficient media sharing for anyone, anytime, and anywhere, technology advances in various aspects need to be re-examined and jointly optimized under a coherent framework. In this proposed long-term research, we will address the above challenges in the latest computing and communication environments. We will focus not only on the optimization of individual components, ranging from media content generation, network and cloud resource allocation, social media propagation, to energy-efficient media computation and communications in cloud and in local devices, but also on their interactions that pose additional problems to designers. A thorough understanding of the modules in this complex system as well as their interactions will facilitate the development of such essential services as social media, IPTV, movie-on-demand, video conferencing, and interactive online gaming in the emerging digital entertainment/business world.We expect six PhD and nine MSc students be trained throughout the 5-year program. These HQPs, together with the technology innovations from our research, will help with maintaining Canada’s lead in the development and deployment of digital and social media services.

在过去的十年中，我们见证了网络多媒体处理和共享向新一代社交、移动和云计算的快速发展。如今，高清视频甚至3D/多视图视频可以通过个人计算设备轻松捕获和浏览，并通过远程云资源方便地处理和存储。用户现在积极参与成为社交生态系统的一部分，而不是被动地接收媒体内容。 3G/4G 无线网络和智能移动设备的深入渗透进一步推动了这场革命，这些设备与在线社交网络和媒体共享服务无缝集成。尽管带宽和计算资源日益丰富，但用户生成媒体内容的数据量不断增加以及社交化共享的无限制覆盖，给内容和网络服务提供商带来了前所未有的挑战。高度多样化的媒体内容类型、来源、分发渠道进一步加剧了网络共享系统中不同组件之间的复杂交互。因此，我们提出以下关键问题：如何通过最新的社交、移动、云计算和通信平台高效地生成海量媒体内容并在海量客户之间分发？除了可扩展性之外，能源也成为这些新一代计算和通信平台的一个关键问题。在移动终端方面，高度受限的电池库长期以来一直是充电间隔时间的瓶颈；在云方面，众所周知，数据中心非常耗电。大量使用CPU、GPU、存储和网络资源的计算和带宽密集型多媒体服务无疑会使情况变得更糟。据预测，多媒体能源消耗将以每年16%的速度增长，占这些服务年度运营支出的一半。为了实现任何人、任何时间、任何地点的可扩展且节能的媒体共享的最终目标，需要在一个连贯的框架下重新审视和联合优化各个方面的技术进步。在这项拟议的长期研究中，我们将在最新的计算和通信环境中解决上述挑战。我们不仅关注各个组件的优化，从媒体内容生成、网络和云资源分配、社交媒体传播，到云和本地设备中的节能媒体计算和通信，而且还关注它们的交互，这些交互给设计人员带来了额外的问题。彻底了解这个复杂系统中的模块及其交互将有助于新兴数字娱乐/商业世界中社交媒体、IPTV、电影点播、视频会议和交互式在线游戏等基本服务的开发。我们预计在整个 5 年计划中将培训 6 名博士生和 9 名硕士生。这些总部项目以及我们研究的技术创新将有助于保持加拿大在数字和社交媒体服务的开发和部署方面的领先地位。